‘‘In-bone’’ Utricle CulturesVA Simplified, AtraumaticTechnique for In Situ Cultures of the Adult Mouse
(Mus musculus) Utricle
*†§Henry C. Ou, *‡Vincent Lin, and *†Edwin W. Rubel
*Virginia Merrill Bloedel Hearing Research Center, ÞDepartment of OtolaryngologyYHead and Neck Surgery,University of Washington, Seattle, Washington, U.S.A.; þSunnybrook Health Sciences Centre, Department
of OtolaryngologyYHead and Neck Surgery, University of Toronto, Toronto, Ontario, Canada; and§Seattle Children’s Hospital, Seattle, Washington, U.S.A.
Hypothesis: The ‘‘in-bone’’ method of culturing utricles de-scribed here is a reliable and atraumatic technique for cultur-ing mature mouse hair cells and studying hair cell death andprotection.Background: The current in vitro technique for studying haircells of the mature mouse utricle involves removal from thetemporal bone and free floating culture in media. This techniquecan be problematic because of variability in the preservationof the sensory epithelium and a steep learning curve that resultsin injury of the sensory epithelium in less experienced hands.We present a new atraumatic technique of culturing the utriclein situ within the temporal bone.Methods: Leaving the temporal bone largely intact, a window isopened in the bony vestibule overlying the mouse utricle. Theentire temporal bone is then placed into culture media. Utricleswere cultured in situ for several days with minimal damage to
the epithelium. The utricles are then fixed in situ, removed from thetemporal bone, and processed. A standardized aminoglycoside-induced hair cell damage protocol was developed.Results:Mature mouse utricles maintained hair cell numbers for3 days in culture. Exposure to neomycin resulted in significantdose-dependent hair cell toxicity (p G 0.0001, 1-way analysis ofvariance). Exposure to the protective drug tacrine resulted in sig-nificant protection against neomycin (p G 0.05, 3-way analysisof variance).Conclusion: The ‘‘in-bone’’ technique is a reliable and atrau-matic method for culturing mature mouse utricles and studyinghair cell death and protection. It is easily mastered and can makein vitro study of hair cells accessible to more research groups.Key Words: Hair cellsVMouseVUtricle.
Otol Neurotol 34:353Y359, 2013.
Hair cell death and protection are frequently studiedwith in vitro preparations. In neonatal mice, this can bedone with whole organ of Corti cultures, because of theimproved survival of neonatal hair cells and ease of dis-section. In contrast, mature mouse inner ear hair cellshave been a challenge to maintain in culture. For thisreason, free-floating vestibular macular cultures have beena useful tool for in vitro evaluation of the mature mousehair cell. This technique has been used effectively in stud-ies of hair cell regeneration and hair cell death, as well asvarious agents that protect hair cells (1Y8).
The utility of the free-floating utricle preparation hasbeen limited, however, by the technical difficulty of the
dissection and the fragility of the fresh tissue when ma-nipulations are required. Frequently, particularly in lessexperienced hands, there is mechanical damage to thesensory epithelium, leading to unusable tissue and morewastage of mice. The steep learning curve involved canmake this technique cumbersome, particularly for newresearchers seeking to gain experience with the inner ear,as well as students with less time to achieve technicalmastery of a technique. In addition, it is inevitable that theprocess of avulsing the utricle from its neural connectionsmay lead to intracellular events that predispose a hair cellto death, or otherwise alter the conditions of the tissue.
This report describes a simple atraumatic method forpreparing mature mouse utricles for culture. The ‘‘in-bone’’method was then used to directly compare neomycin-induced hair cell damage in this new preparation to thefree-floating utricle method. Lastly, protection againstneomycin-induced hair cell death by the protective drugtacrine was demonstrated using the in-bone method.
Address correspondence and reprint requests to Henry C. Ou, M.D.,University of Washington, VMB Hearing Research Center, Box 357923,Seattle, WA 98195-7293; E-mail: [email protected] authors disclose no conflicts of interest.This research was supported by grants from NIH/NIDCD: K08DC009631,
P30-DC004661, and R01-DC005987-05.
Otology & Neurotology34:353Y359 � 2013, Otology & Neurotology, Inc.
AnimalsCBA/CaJ male mice, 4 to 6 weeks old, were obtained from
Harlan Sprague Dawley, Inc. (Indianapolis, IN, USA) and main-tained in the University of Washington Animal Care Facility. Allprocedures described have been approved by the University ofWashington Animal Care and Use Committee.
DissectionMice were sacrificed by cervical dislocation. Each temporal
bone is harvested manually with removal of the bulla and thenplaced into cold sterile culture medium (DMEM, 1% fetal bovineserum, 0.001% ofloxacin). Attached muscle fibers are removed,but neural tissue within the internal acoustic meatus is left intact(Fig. 1A). A small window in the thin bone overlying the utricle(easily identified through the transparent bone due to the whiteotoconia) on the cranial side of the temporal bone is carefullyremoved with a fine bone pick. Care is taken to leave the petrousridge intact. Breaking the petrous ridge can destabilize the bonystructure and lead to separation of the cochlea from the vestibu-lar organs. Adequate bone over the vestibule is removed to exposethe sensory epithelial surface of the utricle (Fig. 1B). The pig-mented membranous labyrinth overlying the sensory epithelium(easily visible in pigmented mice) is then carefully removedwith a number 11 scalpel blade and fine No. 55 Dumont for-ceps to expose the utricle to the culture media without disturb-ing the overlying otoconia. At no point is the utricle’s epithelialor neural side actually touched by any dissecting instruments.Once the pigmented epithelium is opened, the dissection iscomplete, and the temporal bone is placed into a 48-well cultureplate with 1 ml of culture medium at room temperature.Cultures were maintained in a 48-well culture plate for 0 to
3 days in a 37-C incubator, maintained at 5% CO2, changingculture media daily while nutating and rotating at 30 revolutionsper minute, continuously. No neurotrophic factors were addedto the culture media. Leaving the utricle within the temporalbone allows easy and atraumatic manipulation of the tissue aswell as easy changing of culture media without fear of acci-dental aspiration of the utricle. Before fixation, otoconia areremoved; a 25-gauge needle on a 3 ml syringe is used to gentlyblow off otoconia with a steady stream of culture media, whereasthe utricle remains attached within the temporal bone. Again,leaving the utricle within the temporal bone during otoconia re-moval allows for much easier manipulation of the utricle with-out directly contacting the sensory epithelium. Tissue fixation isthen performed by placing the temporal bone overnight in 4%paraformaldehyde at 4-C on a nutator. The fixed temporal boneis then rinsed in phosphate-buffered saline (PBS).Only after fixation is the utricle actually handled with in-
struments. At this point, the utricle can be dissected free withmuch less risk of unintentionally damaging the sensory epithe-lium. Postfixation utricles are harvested from the temporal boneby carefully severing all connections to the membranous labyrinthand vestibular nerve and then grasping the utricle carefully on theneural side and removing it with fine No. 55 Dumont forceps.
Use of Ofloxacin in MediaInitial attempts at whole temporal bone cultures were un-
successful because of high rates of infection. Culture and sen-sitivity analysis by the University of Washington microbiologylaboratory identified several bacteria, including Staphylococcusaureus, Enterobacter cloacae, Acinetobacter baumanni, andProteus mirabilis, all of which were sensitive to members of the
fluoroquinolone family of antibiotics. To combat these bacteria,the fluoroquinolone ofloxacin (Sigma, St. Louis, MO, USA)was prepared as a 0.3% stock solution in PBS and added toculture media at 4 Kl/ml for a final concentration of .001%ofloxacin. We found that ofloxacin is only required during thefirst 24 hours of culture to prevent infection for the duration ofthe culture period. We also tested whether 0.001% ofloxacinaffected hair cell survival in free-floating utricle cultures andfound no difference in survival.
Neomycin Damage ProtocolIn the current study, utricles from in-bone preparations were
cultured in 0, 1, 2, and 4 mM neomycin (Sigma) for 24 hoursand then either fixed overnight or allowed to recover for
FIG. 1. Photomicrograph of mouse temporal bone before (A)and after (B) opening of window in bone overlying utricle. In A, thecircle indicates location for bone removal overlying utricle directlyadjacent to the petrous ridge. In B, the bone overlying the utriclehas been removed with a fine bone pick or No. 11 scalpel blade,and the pigmented epithelium overlying the otoconia has beenremoved with fine forceps without touching the utricle or otoconia.The otoconia are undisturbed and visible overlying the utricularsensory epithelium. The petrous ridge is left intact as it helpsmaintain bony stability. The whole temporal bone is then placeddirectly into the culture well. The utricle sensory epithelium is notmanipulated with any instruments until after the culture period andafter fixation.
48 additional hours in DMEM/1% FBS before fixation in 4%paraformaldehyde for immunohistochemical staining.
Tacrine Protection ProtocolPrevious studies have shown that tacrine exposure has a pro-
tective effect against neomycin exposure in free-floating utricles(8). We sought to determine whether similar results would beobtained using the in-bone method. In-bone utricles were ex-posed to 1 KM tacrine (Sigma) for 4 hours, followed by treat-ment with 0, 2, or 4 mM neomycin for 24 hours, with tacrine stillpresent. After 24 hours of treatment, culture media was replacedwith fresh media (DMEM, 1% fetal bovine serum) without neo-mycin or tacrine. Utricles were left in culture for an additional48 hours after removal of tacrine and neomycin. Five utricleswere used per experimental condition.
Hair Cell CountsOtoconia were removed and utricles were fixed overnight
in 4% paraformaldehyde, as described previously. After rins-ing in PBS, the utricles were placed in blocking solution (2%bovine serum albumin, 0.4% normal goat serum, 0.4% normalhorse serum, and 0.4% Triton X-100 in PBS) for 3 hours at roomtemperature. Utricles were then double labeled using antibodiesagainst calmodulin (monoclonal mouse, 1:200; Sigma) andcalbindin (polyclonal-rabbit, 1:250; Chemicon, Billerica, MA,USA) overnight at 4-C (5). After additional PBS rinses, theutricles were incubated with secondary antibodies (Alexa 594goat antirabbit IgG 1:500 and Alexa 488 horse antimouse IgG1:500; Invitrogen, Carlsbad, CA, USA) for 2 hours at roomtemperature. Utricles were then washed in PBS and mounted inFluoromount-G (Southern Biotech, Birmingham, AL, USA).Utricles were examined using either a Zeiss Axioplan II fluo-
rescence microscope or an Olympus FV-1000 confocal microscope(Fig. 2). Hair cell counts from 4 randomly selected 900 Km2
areas from the striolar and the extrastriolar regions were deter-mined (Cunningham et al., 2002). For each utricle, hair cells inthe 4 areas were totaled for striolar and extrastriolar regions sepa-rately. These hair cell totals were then averaged for the mean hair
cell density in striolar and extrastriolar regions and are presented asa percentage of control utricles (T standard deviation [SD]) culturedwithout neomycin.
Free-Floating Utricle CulturesFor purposes of comparison to the in-bone method, utricles
were cultured using the free-floating utricle method, as previ-ously described (2,5). Mature mice (4Y6 wk of age) were sac-rificed by cervical dislocation. Utricles were then removed fromthe temporal bones using sterile technique and cultured in tis-sue culture plates with a 2:1 mixture of basal medium Eagle andEarle’s balanced salt solution, with 5% fetal bovine serum(Invitrogen). Utricles were cultured for 24 hours with 0, 1, 2, or4 mM neomycin. Five utricles were cultured per experimentalcondition. After neomycin treatment, otoconia were removedby a steady stream of phosphate-buffered solution (PBS), andthen, utricles were fixed overnight in 4% paraformaldehyde at4-C. After this, tissue was processed exactly as above. Utricleswere also cultured for 24 hours with 0, 1, 2, or 4 mM neomy-cin in the presence of 0.001% ofloxacin to determine whetherthere was any effect of ofloxacin on hair cell survival in free-floating cultures.
Scanning Electron Microscopy of In-bone UtriclesUtricles prepared with the in-bone method were cultured for
1 hour or 72 hours in DMEM/1% FBS and then rinsed in PBSbefore fixation in 4% glutaraldehyde in 0.1 M sodium phosphateat 4-C overnight. Specimens were then rinsed 3 times in 0.1 Msodium phosphate buffer (PB) and then postfixed in 1% osmiumtetroxide in 0.1 M PB for 30 minutes in an ice bath. Specimenswere then rinsed in 0.1 M PB and dehydrated through a gradedethanol series: 35%, 70%, 95%, and 100% (�2). Utricles werethen critical point dried, mounted on SEM stubs, and sputtercoated with Au/Pd. Note that utricles were kept in situ within thetemporal bone through processing and imaging. SEM was per-formed using a JEOL JSM-840A scanning electron microscope.
StatisticsAll values were calculated and presented as the mean value T
1 SD. Statistical analyses were performed using 1-way ANOVA(Vassarstats, http://faculty.vassar.edu/lowry/VassarStats.html) and3-way ANOVA (PASW Statistics 18.0; SPSS, Inc., Chicago,IL, USA). Results were considered statistically significant ifp G 0.05.
RESULTS
Utricular Hair Cell Survival in Culture MediaAdult utricles were cultured for 1 to 3 days in culture
media. After 3 days in culture media, hair cell numberswere well preserved (Fig. 2). Hair cells were then countedfrom striolar and extrastriolar regions. Mean hair cell den-sity per 1,000 Km2 (SD) in striolar and extrastriolar areaswas 11.0 (4.4) and 11.7 (1.9), respectively, after 24 hoursof culture. After 3 days in culture media, mean hair celldensity per 1,000 Km2 in striolar and extrastriolar areaswas 11.5 (3.1) and 14.7 (2.5), respectively, demonstratingfavorable stability of hair cell density while in this cul-ture condition (n = 5Y9 utricles per group). Scanning elec-tron microscopy of in-bone utricles fixed after 72 hoursin culture demonstrated excellent stereocilia morphologywhen compared with utricles fixed after 1 hour in culture
FIG. 2. Confocal image of mouse utricle cultured for 3 days andthen fixed and labeled with antibodies against calmodulin (green)and calbindin (red) to delineate striolar (S) and extrastriolar (ES)hair cells. Hair cells are well preserved. Scale bar = 50 Km.
(Fig. 3). Many hair cells exhibited blebbing of the apicalplasma membrane after 1 hour in culture media. This seemedreduced after 72 hours in culture media, presumably be-cause of acclimatization of hair cells to the media.
Neomycin Damage ProtocolUtricles were cultured for 24 hours with 0, 1, 2, and
4 mM neomycin using the in-bone method and then fixedfor fluorescence microscopy. The mean (T1 SD) striolarand extrastriolar hair cell numbers expressed as a per-centage of undamaged (no neomycin) controls are shownin Figure 4A. Hair cell counts after 24 hours in culturemedia with neomycin showed no significant hair cell lossin striolar or extrastriolar regions (striolar, p 9 0.20 value;extrastriolar, p 9 0.20, 1-way ANOVA). This is in con-trast to marked dose dependent hair cell loss demon-strated with free floating utricles exposed to 24 hours of1, 2, and 4 mM neomycin (Fig. 4A; striolar, p G 0.0001;extrastriolar, p G 0.0001; 1-way ANOVA). In addition, todetermine whether ofloxacin was having a protective ef-fect on hair cells, we also cultured free-floating utriclesexposed to 0, 1, 2, and 4 mM neomycin with and without0.001% ofloxacin for 24 hours and found no differencein hair cell survival (p 9 0.05, 2-way ANOVA, data notshown).
Because no hair cell death was seen with the in-bonemethod at 24 hours, we examined whether hair cell death
would emerge over a longer period. After rinsing awayneomycin, utricles were cultured for 48 additional hours(24 h with neomycin followed by 48 hours in culturemedia without neomycin). Immunohistochemistry was per-formed to measure hair cell counts in striolar and extra-striolar regions. With this 72-hour protocol, there wassignificant dose-dependent hair cell loss. Figure 5B showsan example of a utricle cultured by this protocol after 4 mMneomycin compared with a utricle cultured for the sameperiod but without neomycin exposure (Fig. 5A). Bothstriolar and extrastriolar hair cells in the in-bone cultureswere dramatically affected (Fig. 4B; p G 0.001 for bothstriolar and extrastriolar hair cells, 1-way ANOVA). In-terestingly, although at 24 hours, the 2 methods yieldeddissimilar results, by 72 hours the in-bone damage protocoldata closely matches with damage seen with free floatingmouse utricles cultured for 24 hours with neomycin.
Tacrine-Induced Hair Cell ProtectionWe have previously identified tacrine as a drug that
protects zebrafish lateral line hair cells and hair cellsfrom cultured explanted utricles of mature mice againstneomycin-induced hair cell death (8). To determine whetherprotective drugs found effective in our zebrafish screen andwith free floating utricles were also protective with the in-bone method, hair cell counts were performed on utri-cles exposed to 2 and 4 mM neomycin with and without
FIG. 3. Scanning electron micrographs of mouse utricles cultured using in-bone method. A and B demonstrate in-bone utricle after 1 hourin culture media. In A, the utricle sensory epithelium is easily visualized within the bony vestibule. At higher magnification (B), the stereociliabundles are visible. Small blebs of the apical plasma membrane are seen and are likely because of the acute exposure to culture media.C and D demonstrate in-bone utricle after 72 hrs in culture media. After 72 hours in culture (C), the utricle sensory epithelium remains wellvisualized in the bony vestibule with excellent morphology of stereocilia seen at higher magnification (D). Blebbing has also decreased,potentially because of acclimatization of hair cells to culture conditions after 72 hours.
tacrine (Fig. 6). Tacrine pretreatment increased hair cellsurvival after 4 mM from 17% (mean hair cell density,1.06 per 1,000 Km2; no tacrine) to 46% (mean hair celldensity 4.31 per 1,000 Km2; with tacrine) in striolarregions of the utricle. In extrastriolar regions, tacrine pre-treatment increased hair cell survival after 4 mM neomy-cin from 35% (mean hair cell density, 4.44 per 1,000 Km2;no tacrine) to 57% (mean hair cell density 7.19 per
1,000 Km2; with tacrine). This protective effect of tacrinewas significant against both 2 and 4 mM neomycin inboth striolar and extrastriolar regions (p G 0.05, repeatedmeasures 3-way ANOVA).
DISCUSSION
Viable culture preparations from adult mammalian in-ner ear have been problematic. There is currently no satis-factory way to culture mature organ of Corti for a prolongedduration. As a result, most in vitro studies using adult in-ner ear tissue have avoided using cochlea, instead usingfree floating utricle explants, which survive nicely in cul-ture (1,2,4,9). This technique, although effective, is tech-nically difficult because of the size and fragility of thesensory epithelium.
FIG. 4. Neomycin dose-response curves showing hair cell sur-vival in utricles cultured for 24 hours (A) and 72 hours (B). A,Striolar and extrastriolar hair cell survival after 24 hours in vitro iscompared between free-floating and in-bone utricle cultures.Utricles were exposed to 24 hours of neomycin at 0-, 2-, and 4-mMconcentrations and then fixed and analyzed. B, In-bone cultureswere exposed to neomycin for 24 hours and then rinsed and thencultured in fresh media (without neomycin) for an additional48 hours. Hair cell loss for in-bone cultures maintained for a total of72 hours are compared with data from free floating cultures, re-peated from A. In A, in-bone striolar (solid line •) and in-boneextrastriolar (solid line h;) there was no hair cell loss in response toneomycin. This is in comparison to the dose-response curves forstriolar (dotted line 0) and extrastriolar (dashed line r) hair cells infree floating utricles exposed to identical neomycin treatments. In B,after 24 hours of neomycin exposure, and 48 additional hours inculture without neomycin, in-bone utricles demonstrate increasingstriolar (solid line •) and extrastriolar (solid line h;) hair cell loss withincreasing neomycin concentrations, more closely resembling thefree floating utricle cultures after just 24 hours (dotted line 0 anddashed line r). Data points represent the mean of 5 utricles for eachexperimental condition. Error bars represent the SD from the mean.
FIG. 5. Immunocytochemistry of neomycin-induced hair celldeath in the mouse utricle cultured using the in-bone technique.Utricles were cultured for 24 hours with 0 (A) and 4 mM (B) neo-mycin, and then rinsed several times to remove neomycin andallowed to recover for 48 additional hours. Utricles were then fixedand labeled with antibodies against calmodulin (green) and cal-bindin (red). In A, there is excellent survival of striolar (red) andextrastriolar hair cells. In B, with 4-mM neomycin, there is markedstriolar and extrastriolar hair cell loss. Scale bar in B = 50 Km.
The in-bone culture method we discuss here allows theadult mouse utricle to be harvested and left in culture for3 days. Because the utricle is left within the temporalbone and not manipulated until after fixation, there ismuch less tissue damage and, thus, much more consistenttissue quality. In addition, removal of otoconia is signif-icantly easier with the utricle left attached within the tem-poral bone. This alone makes this technique much moreaccessible and more easily taught to students and technicalstaff. In our experience, achieving expertise in the tradi-tional utricle explant technique often can require monthsof practice, whereas the in-bone method can be masteredwithin days.
Aminoglycoside Damage Occurs PredictablyAs with the free floating utricle cultures, aminoglycoside
damage with the in-bone method occurs with a predictabledose-response relationship. However, it is interesting tonote that damage occurs at a slower rate. With the in-bonemethod and neomycin exposure up to 4 mM for 24 hours,hair cell loss was not seen immediately but was very evi-dent 2 days later, even when no neomycin was presentduring the final 48 hours of incubation. On the other hand,significant hair cell loss occurs after 1 day with free float-ing utricles in our study as well as others (5,6). Whetherthe in-bone culture method or the free floating methodmore closely resembles in vivo exposure conditions re-mains unclear. It should be noted that the time course ofutricular hair cell loss after aminoglycoside exposure invivo is not well described. Although most in vivo amino-glycoside damage protocols seem to require several daysto cause damage (10), more recent studies demonstrate that
aminoglycosides can be detected in the inner ear withinhours after systemic administration (11,12). In addition,others have found that hair cell death occurs within hoursafter direct exposure of hair cells to aminoglycosides invivo in guinea pig (12). It should be noted that in the in-bone culture, aminoglycosides would only have accessto the apical side of the utricular hair cells, whereas freefloating cultures, although still protected by otoconia, arecircumferentially bathed in solutions containing amino-glycoside, potentially increasing the toxicity. Althoughunproven, one could also hypothesize that the in-bone cul-ture method, by leaving neural connections relatively intactand sensory epithelium undisturbed, might result in haircells that are less fragile than those in free floating utriclesand thus less susceptible to aminoglycoside damage. Thesedifferences may account for some of the differences seenbetween the in-bone and free floating utricle cultures.
Difficulties With In-bone MethodInitial attempts at culturing whole temporal bones were
hindered by difficulties with culture media infection. Mi-crobiologic testing of the infected media yielded multipleorganisms, all of which were sensitive to fluoroquinoloneantibiotics. Cultures thereafter were thus cultured with0.001% ofloxacin with excellent results and no apparentototoxicity based on hair cell counts. It is notable that ac-cording to current recommendations from the Ameri-can Academy of OtolaryngologyYHead and Neck Surgery(13,14), only fluoroquinolone antibiotics are consideredsafe to use in the middle ear and ‘‘less’’ ototoxic. Neverthe-less, to minimize any possible effect that ofloxacin mighthave on hair cell survival, culture media was changed after24 hours to remove ofloxacin from the preparation. Thisinitial 24-hour exposure to 0.001% ofloxacin was suffi-cient to prevent any subsequent infection during 5-dayculture periods. In addition, to rule out the possibility thatofloxacin was protecting hair cells against neomycin, weexposed free floating utricles to neomycin with and with-out ofloxacin present and found no significant effect.
CaveatsThe methods described here are meant to serve as an
alternative to the free-floating preparation of the adultmouse utricle. Although there are advantages in terms oftechnical ease, difficulties with infection do require theaddition of an antibiotic (ofloxacin) to the culture media.In addition, leaving the utricle within the temporal bone mayexpose cultured hair cells to additional survival factors gen-erated by other tissues (bone, muscle, and nerves) in theculture media. This ‘‘co-culture’’ scenario could introducemore variability into the system, although in this study,the results from damaging and protective compounds werequite consistent. Clearly, the free-floating utricle prepara-tion, as well as isolated epithelium preparations (15), re-mains valuable tools for investigating many questions abouthair cell viability and regeneration.
Acknowledgments: The authors thank Dale Cunningham forexcellent technical expertise in electron microscopy.
FIG. 6. Pretreatment with tacrine protects against neomycin-induced hair cell death in in-bone utricle cultures. Utricles werecultured using the in-bone method and then pretreated either with0.01% DMSO control or 1 mM tacrine in culture media. Utricleswere then treated with 2 or 4 mM neomycin for 24 hours followedby 48 hours in culture media alone without neomycin or tacrine.Pretreatment with tacrine significantly protected striolar and ex-trastriolar hair cells against 2 and 4 mM neomycin when comparedwith DMSO pretreatment controls (*p G 0.05, repeated measures,6-way ANOVA). Data bars represent the mean of 5 utricles (blackbars = extrastriolar hair cells, hatched bars = striolar hair cells).Error bars represent the SD from the mean.
1. Warchol ME, Lambert PR, Goldstein BJ, Forge A, Corwin JT.Regenerative proliferation in inner ear sensory epithelia from adultguinea pigs and humans. Science 1993;259:1619Y22.
2. Oesterle EC, Tsue TT, Reh TA, Rubel EW. Hair-cell regenerationin organ cultures of the postnatal chicken inner ear. Hear Res 1993;70:85Y108.
3. Lambert PR. Inner ear hair cell regeneration in a mammal: identifi-cation of a triggering factor. Laryngoscope 1994;104(6 Pt 1):701Y18.
4. Yamashita H, Oesterle EC. Induction of cell proliferation in mam-malian inner-ear sensory epithelia by transforming growth factor alphaand epidermal growth factor. Proc Natl Acad Sci U S A 1995;92:3152Y5.
5. Cunningham LL, Cheng AG, Rubel EW. Caspase activation in haircells of the mouse utricle exposed to neomycin. J Neurosci 2002;22:8532Y40.
6. Cunningham LL. The adult mouse utricle as an in vitro preparationfor studies of ototoxic-drug-induced sensory hair cell death. BrainRes 2006;1091:277Y81.
7. Owens KN, Santos F, Roberts B, et al. Identification of genetic andchemical modulators of zebrafish mechanosensory hair cell death.PLoS Genet 2008;4:e1000020.
8. Ou HC, Cunningham LL, Francis SP, et al. Identification of FDA-approved drugs and bioactives that protect hair cells in the zebrafish
(Danio rerio) lateral line and mouse (Mus musculus) utricle. J AssocRes Otolaryngol 2009;10:191Y203.
9. Oesterle EC, Rubel EW. Hair cell generation in vestibular sensoryreceptor epithelia. Ann N Y Acad Sci 1996;781:34Y6.
10. WuWJ, Sha SH, McLaren JD, Kawamoto K, Raphael Y, Schacht J.Aminoglycoside ototoxicity in adult CBA, C57BL and BALB miceand the Sprague-Dawley rat. Hear Res 2001;158:165Y78.
11. Imamura S, Adams JC. Distribution of gentamicin in the guinea piginner ear after local or systemic application. J Assoc Res Otolaryngol2003;4:176Y95.
12. Wang Q, Steyger PS. Trafficking of systemic fluorescent genta-micin into the cochlea and hair cells. J Assoc Res Otolaryngol 2009;10:205Y19.
13. Matz G, Rybak L, Roland PS, et al. Ototoxicity of ototopical anti-biotic drops in humans. Otolaryngol Head Neck Surg 2004;130:S79Y2.
14. Roland PS, Stewart MG, Hannley M, et al. Consensus panel on roleof potentially ototoxic antibiotics for topical middle ear use: Intro-duction, methodology, and recommendations. Otolaryngol HeadNeck Surg 2004;130:S51Y6.
15. Saffer LD, Gu R, Corwin JT. An RT-PCR analysis of mRNA forgrowth factor receptors in damaged and control sensory epitheliaof rat utricles. Hear Res 1996;94:14Y23.
